Computational Design of Unnatural Amino Acid Dependent Metalloproteins

描述（由申请人提供）：非天然氨基酸的位点特异性结合和计算蛋白设计领域的融合代表了目前尚未探索但有前途的生化研究途径。虽然计算方法已经开发用于天然存在的蛋白质，但用这些技术处理非天然氨基酸的能力尚未得到充分探索。该研究旨在开发一种计算方法，允许设计含有非天然氨基酸的蛋白质，最终目标是产生具有治疗潜力的新型非天然氨基酸依赖酶。由华盛顿大学贝克实验室成员开发的罗塞塔软件套件将首先用于设计铁结合蛋白，该蛋白利用金属结合的非天然氨基酸联吡啶丙氨酸——舒尔茨和斯克里普斯研究所的同事首先将其结合到蛋白质中。由于这种非天然氨基酸对铁具有固有的亲和力，因此在蛋白质中设计金属结合位点的难题应该在计算上更容易处理。第二个目标是同时设计多巴胺的结合位点（它将为铁提供两种氧配体）。像多巴胺这样的儿茶酚对铁具有固有的高亲和力，这表明工程蛋白可以作为这类重要小分子的传感器。最后，儿茶酚结合蛋白将进一步通过计算设计，目标是创造一种非天然氨基酸依赖的外二醇双加氧酶。这种酶可能对持久性拟人化毒素（如多氯联苯化合物）的生物修复产生深远影响。设计的含有蛋白质的非天然氨基酸将在细菌表达系统中使用舒尔茨实验室成员开发的技术生产。然后，纯化的蛋白质将使用一系列生物分析技术进行分析，这些技术将根据特定的目的检查金属或儿茶酚的结合能力，或酶的活性。在实验过程中收集的数据将用于未来设计本项目范围内外的其他含蛋白质的非天然氨基酸。因此，这项研究应该在生物科学领域产生深远的影响，将超出上述项目的范围。由于本提案中探索的两个科学领域目前都处于快速发展的状态，在本研究过程中收集的任何信息都将指导进一步的计算设计工作，包括其他目前可用的、遗传编码的非天然氨基酸，以及未来开发的氨基酸。

DESCRIPTION (provided by applicant): The confluence of the fields of site-specific incorporation of unnatural amino acids and computational protein design represents a currently unexplored but promising avenue of biochemical research. While computational methods have been developed for naturally occurring proteins, the ability to treat non-natural amino acids with these techniques has yet to be fully explored. The research proposed seeks to develop a computational method that allows design of proteins containing unnatural amino acids, with the ultimate goal of generating novel unnatural amino acid dependent enzymes with therapeutic potential. The Rosetta suite of software developed by members of the Baker lab at the University of Washington will first be used to design iron binding proteins that utilize the metal binding unnatural amino acid bipyridyl alanine - first incorporated into proteins by Schultz and co-workers at The Scripps Research Institute. As this unnatural amino acid has inherent affinity for iron, the difficult problem of designing a metal binding site within a protein should be rendered more computationally tractable. As a second goal, a binding site for dopamine (which will provide two oxygen ligands for the iron) will be concurrently engineered. Catechols like dopamine have inherently high affinities for iron suggesting the engineered proteins could serve as sensors for this important class of small molecules. Finally, the catechol binding proteins will be further designed computationally with the goal of creating a non-natural amino acid dependent extradiol dioxygenase like enzyme. Such an enzyme could have a far-reaching impact with respect to bioremediation of persistent anthropomorphic toxins such as polychlorinated biphenyl compounds. The designed unnatural amino acid containing proteins will be produced in a bacterial expression system using techniques developed by members of the Schultz laboratory. Purified proteins will then be analyzed using a host of bioanalytical techniques that will examine metal or catechol binding abilities, or enzymatic activity depending on the specific aim. Data collected in the course of experimentation will be used for future design of other unnatural amino acid containing proteins both within the scope of this project, and beyond. Consequently, this research should have far reaching impacts within the biological sciences that will extend beyond the projects described above. As both of the scientific fields explored in this proposal are currently in a state of rapid growth, any information gleaned in the course of this research will guide further computational design efforts involving other currently available, genetically encoded non-natural amino acids, as well as those developed in the future. PUBLIC HEALTH RELEVANCE: This research ultimately seeks to engineer unnatural amino acid containing proteins that possess the ability to catalytically degrade polychlorinated biphenyl environmental toxins. Additionally, the computational methods developed in the course of the research will provide vital information that will guide future efforts for the design of unnatural amino acid containing proteins with therapeutic and other useful functions.